How to Assess PV Trackers vs Fixed Tilt: What the US Plant Data Show on Capacity Factor and Cost, Where Fixed Tilt Is Still Chosen, the IEC 62817 and GB/T 29320 Tracker Standards, and How Models Treat Backtracking

Short answer. The choice is a yield-versus-cost-and-site trade. On yield, LBNL's US plant data show single-axis tracking adding about 5 percentage points to median capacity factor over fixed tilt, more where the solar resource is strong, within an overall range of 17% to 31% by region. On cost, the historical premium for trackers has fluctuated and at times reversed; in LBNL's preliminary 2024 sample tracking projects cost less ($1.61/WAC) than fixed-tilt projects ($1.90/WAC), though the comparison does not control for other project differences, while NREL's bottom-up Q1 2020 benchmark put one-axis tracking at $1.01/WDC against $0.94/WDC for fixed tilt. On site, EIA explains why: tracking is effective for direct-beam radiation but gives no benefit for diffuse radiation, so cloudier regions gain less, and fixed tilt is cheaper to install and needs less land. The outcome in the US is that 99% of new utility-scale capacity in 2024 used single-axis tracking and fixed tilt survives only in edge cases such as challenging terrain or high wind loading. For the hardware itself, IEC 62817 is the design-qualification standard for trackers and China's GB/T 29320-2024 sets technical requirements for tracking systems in PV plants. All yield and cost figures below are US data; no Chinese figure is quoted because no retrieved official source states one.

1. Why tracking helps, and where it does not

EIA's 2017 analysis explains the physics in plain terms: tracking is effective at capturing direct-beam radiation because the panel can be aimed at the sun, but it gives no benefit for diffuse radiation, which arrives from many directions; the cloudier eastern United States therefore receives a higher share of diffuse radiation, which reduces the benefit of tracking and explains the higher share of fixed-tilt systems there. Fixed-tilt units are cheaper to install, while single- and dual-axis trackers need more land and cost more but produce more. EIA cites NREL's PVWatts estimates for a 10-kW system in Los Angeles: single-axis tracking produces 21% more electricity a year than a fixed 20-degree tilt, and dual-axis 31% more.

The regional split in EIA's end-2015 data followed that logic: more than 50% of operating US utility-scale PV systems used some form of tracking, located mainly in the Southwest; east of the Mississippi fixed tilt was 80% (2,468 MW) of utility-scale PV capacity and single-axis only 19%, while west of it fixed tilt was 37% (3,904 MW) and single-axis 56%.

2. What the plant data show on capacity factor

LBNL's 2025 data update, built on US utility-scale plant records, finds that average capacity factors range from 17% in the least-sunny regions to 31% in the sunniest; that tracking adds about 5 percentage points to median capacity factor versus fixed tilt, with the improvement more pronounced in higher-resource areas and above five points in the strongest-resource regions; that the strongest-resource quartile has a 9-point higher median capacity factor than the lowest; and that the highest inverter-loading-ratio quartiles add about 1 point.

NREL's Q1 2020 benchmark used the same ordering in its LCOE inputs: 21.6% (fixed tilt) against 25.2% (one-axis) in Phoenix, 17.3% against 19.6% in Kansas City, and 16.2% against 18.1% in New York City. CNWattHub arithmetic on those pairs: the tracking gain is 3.6 points in the high-resource case, 2.3 in the medium case and 1.9 in the low case, consistent with LBNL's observation that the benefit grows with resource.

NREL's 2025 ATB treats one-axis tracking as the reference utility-scale design (100 MWDC / 75.8 MWAC, bifacial modules, inverter loading ratio 1.32) and states that its capacity factors are for one-axis tracking and do not represent the lower capacity factors of fixed-tilt systems; for US single-axis projects installed 2007–2023 the cumulative median AC capacity factor was 25% with a 7%–35% project range, and its projected future gains include 'better tracking' among other improvements.

3. What the data show on cost

Two US sources give the two sides of the cost picture.

Bottom-up model (NREL, Q1 2020). Utility-scale PV at $0.94/WDC ($1.28/WAC) for fixed tilt and $1.01/WDC ($1.35/WAC) for one-axis tracking; O&M of $16.32/kWDC-yr for fixed tilt and $17.46/kWDC-yr for single-axis tracking; modelled DC-to-AC ratios of 1.37 for fixed tilt and 1.34 for trackers. NREL noted that cumulative US tracking installation reached 65% in 2019 with 82% of new 2019 installations tracking, and therefore used fixed tilt for its 2010–2015 LCOE benchmarks and one-axis tracking for 2016–2020. CNWattHub arithmetic: the modelled tracker premium was $0.07/WDC, about 7% of the fixed-tilt system cost, and $1.14/kWDC-yr of O&M.

Observed plant costs (LBNL, 2024 vintage). Capacity-weighted installed costs of $1.61/WAC ($1.22/WDC) for tracking projects and $1.90/WAC ($1.35/WDC) for fixed-tilt projects. LBNL warns that the comparison does not control for other cost drivers (equipment, labour, land, interconnection, project size), that trackers can bear higher upfront cost because they deliver more energy per installed capacity, that the premium has fluctuated and at times reversed (2016 and 2024), and that the 2024 estimates are preliminary. CNWattHub reading: the 2024 reversal is a sample result, not a rule that trackers are cheaper.

4. Where fixed tilt is still chosen

LBNL reports that 99% of new US utility-scale capacity in 2024 used single-axis tracking, that the upfront premium for trackers faded over time, and that fixed tilt is now chosen only in edge cases such as challenging terrain or high wind loading. Fixed tilt is now built mainly at lower-insolation sites, with exceptions in windy regions (Florida), on brownfields and landfill sites and on particularly challenging terrain; since 2013 the tracking share of US utility-scale PV grew from 53% to 95%. CNWattHub reading for a project screen: in LBNL's account, terrain, wind load, site type and resource are what keep fixed tilt in play.

5. The standards

IEC 62817. IEC 62817:2014 is a design-qualification standard for solar trackers in PV systems (usable for other solar applications). It defines test procedures for key components and for the complete tracker; some procedures produce parameters reported in a defined tracker specification sheet, others give a pass/fail result intended to separate designs likely to fail early from sound designs. The IEC webstore lists a consolidated version IEC 62817:2014+AMD1:2017+AMD2:2026, that is the 2014 edition with amendments of 2017 and 2026. A buyer should ask which edition and amendments a tracker's qualification report covers and ask for the specification sheet the standard defines; the amendments' technical content is not described here because the full text was not read.

GB/T 29320-2024. China's recommended national standard 光伏电站太阳跟踪系统技术要求 (Technical requirements for solar tracking system of photovoltaic power station) was published and took effect on 23 August 2024 and is administered by the China Electricity Council. Only its title and dates are cited here.

6. What a yield model needs to know

NREL's SAM documentation defines backtracking as a one-axis tracking strategy that avoids row-to-row shading: without it a tracker points the modules at the sun, so closely spaced rows shade each other at some sun angles; with it the tracker orients the modules away from the sun to avoid the shading. The tracker rotation limit is the maximum and minimum allowable rotation angle (for example 45 degrees either side of horizontal), and terrain slope and azimuth inputs apply only to one-axis tracking systems. CNWattHub reading: these three inputs (backtracking, rotation limit and terrain) apply to the tracker case and should be stated explicitly in any yield comparison.

Reading table

Checklist

CNWattHub editorial method, not a sourced rule:

1. Classify the site by resource and by diffuse share; the tracking gain is largest where direct radiation dominates.
2. Screen for the fixed-tilt edge cases first: terrain, wind loading, brownfield or landfill constraints, land availability.
3. Run both mounting types in the same model with the tracker inputs stated (backtracking on or off, rotation limit, terrain angles) and the DC-to-AC ratio held comparable.
4. Compare installed cost and O&M per kW for both options on the same scope; treat any published premium or discount as sample-specific.
5. Ask the tracker supplier for the IEC 62817 qualification report (edition and amendments) and the tracker specification sheet, and for conformity with GB/T 29320-2024 for Chinese sites.
6. Record the capacity-factor assumption you adopt and its source, because tracking and fixed-tilt figures are not interchangeable.

Scope limits

US data from LBNL, EIA and NREL and two standards' public pages only. No Chinese tracker share, cost or yield gain is quoted; the standards' full texts were not read; LBNL's 2024 cost comparison is preliminary and uncontrolled. Not engineering, financial or investment advice.